Three-dimensional reconstruction method for pathological image features of hyperplastic holescoma

Abstract

The invention provides a hyperplastic sheathing tumor pathological image feature three-dimensional reconstruction method, and relates to the technical field of three-dimensional reconstruction, and the method comprises the steps: obtaining a pathological section image and depth information of the pathological section image, and generating a pathological section depth image containing the depth information; emitting a laser signal, detecting a pathological tissue, and extracting a high feature region of the pathological tissue; emitting gamma rays, detecting the pathological tissue, and generating a gamma ray absorption key area highlighting the tumor area of the pathological tissue; and three-dimensional information of the pathological section depth image is supplemented by combining an extracted high-feature region and a gamma ray absorption key region of the pathological tissue, so that the three-dimensional form of the reconstructed hyperplastic sheathing tumor pathological image is obtained. According to the method, detailed anatomical information is provided for subsequent treatment scheme formulation.

Description

Technical Field

[0001] The present invention provides a three-dimensional reconstruction method for the pathological image features of proliferative trichilemmal tumor, which relates to the technical field of ultrasonic image processing. Background Art

[0002] In the medical field, accurate diagnosis of proliferative trichilemmal tumor is of great importance. Pathological section examination is one of the important ways to diagnose proliferative trichilemmal tumor. In this process, doctors obtain the morphological information of tissues by observing pathological sections under a microscope. However, conventional pathological section images are only two-dimensional plane images, and the information they provide is severely limited. Two-dimensional images cannot show the structure and features of pathological tissues in the depth direction, making it difficult for doctors to comprehensively and stereoscopically grasp the true situation inside the tissues. Key information such as the infiltration range of the tumor at different depths and the positional relationship with surrounding normal tissues in three-dimensional space is difficult to clearly obtain from two-dimensional section images, which greatly affects the accuracy and comprehensiveness of diagnosis. In terms of detecting key feature regions of pathological tissues, previous technologies also have deficiencies. Ordinary optical observation means are difficult to accurately locate high-feature regions in pathological tissues that have special biological or morphological significance. These high-feature regions often contain key clues to important characteristics such as tumor growth, differentiation, and invasion. Due to the lack of effective extraction means, doctors are prone to miss these key information, resulting in inaccurate assessment of tumors. In addition, when highlighting the tumor regions of pathological tissues, existing technologies also face challenges. Some traditional imaging technologies are difficult to clearly outline the boundaries and ranges of deep tumors or tumor regions with low contrast to surrounding tissues. This makes it difficult for doctors to accurately judge the actual size, shape, and invasion degree of tumors, thus affecting the precise formulation of subsequent treatment plans. Therefore, traditional pathological diagnosis technologies have obvious deficiencies in obtaining comprehensive information of pathological tissues, accurately extracting key feature regions, and clearly highlighting tumor regions. Summary of the Invention

[0003] To solve the above technical problems, the present invention provides a three-dimensional reconstruction method for the pathological image features of proliferative trichilemmal tumor, including: obtaining a pathological section image and the depth information of the pathological section image to generate a pathological section depth image containing depth information; emitting laser signals to detect the pathological tissue and extract the high-feature regions of the pathological tissue; emitting gamma rays to detect the pathological tissue and generate a key gamma-ray absorption region highlighting the tumor region of the pathological tissue; combining the high-feature regions extracted from the pathological tissue and the key gamma-ray absorption region and supplementing them into the three-dimensional information of the pathological section depth image to obtain the three-dimensional morphology of the reconstructed pathological image of proliferative trichilemmal tumor.

[0004] In a preferred embodiment, during laser signal acquisition, the pathological tissue absorbs laser energy to generate a transmission signal. Transmission data at different angles are obtained through a detector. The transmission data at different transmission angles are collected to obtain a transmission laser image. The transmission data is projected onto a two-dimensional plane using a mapping function, the feature map of the transmission laser image is calculated, the frequencies of occurrence of each eigenvalue are counted, and the high-feature region is extracted.

[0005] In a preferred embodiment, let the transmission data collected by the detector at the transmission angle θ be P(θ, r), where r is the radial coordinate, and the transmission laser image I is obtained. PA The formula is: ; where h(r - r ， ) is the mapping function, and r ， represents the radial mapping coordinate at the transmission angle θ of the point with coordinates (x, y) in the image; Calculate the feature map of the transmission laser image I PA , count the frequencies of occurrence of each eigenvalue, traverse all set thresholds T, and calculate the difference value : ; where is the proportion of the two types of features after segmentation according to the threshold T; is the average eigenvalue of the two types of features, is the global eigenvalue; Traverse all thresholds T to calculate , and select the T that maximizes as the segmentation threshold. The region with an eigenvalue greater than the segmentation threshold is identified as the high-feature region and extracted.

[0006] In a preferred embodiment, let the projection data formed by the detector recording gamma rays be p m , and use an iterative algorithm to convert the projection data into a tissue absorption distribution image: ; where represents the absorption value of the tissue absorption distribution image corresponding to voxel i at the k-th iteration; represents the absorption value of the tissue absorption distribution image corresponding to voxel i at the (k + 1)-th iteration, A m,i is the system response matrix, which describes the contribution of voxel i to detector unit m, R m,i is the normalization factor, and k is the number of iterations. The tissue absorption distribution image is gradually optimized through multiple iterations.

[0007] In a preferred embodiment, a three-dimensional Gaussian convolution is performed on the tissue absorption distribution image to denoise it, and the denoised absorption value corresponding to voxel i in the tissue absorption distribution image is calculated : ; where Ω is the spatial neighborhood, G is the Gaussian kernel function, where j - i represents the spatial position difference between voxel j and voxel i, and ρ is the smoothing parameter that controls the degree of smoothing; Based on a threshold-based segmentation method, an absorption threshold T is selected H , and the key gamma-ray absorption regions are extracted: ; All regions with a value of 1 form the key gamma-ray absorption regions.

[0008] In a preferred embodiment, DPA q is used to represent the image feature value of the q-th neighborhood voxel in the high-feature region, and SPT q is used to represent the image feature value of the q-th neighborhood voxel in the key gamma-ray absorption region, and the correlation coefficient r between the high-feature region and the key absorption region is calculated DS : ; where Q is the total number of neighborhood voxels, q is the q-th neighborhood voxel, is the average value of the image feature values of all neighborhood voxels in the high-feature region, is the average value of the image feature values of all neighborhood voxels in the key absorption region; Based on the correlation coefficient r DS a feature-absorption correlation map is generated: ; where I G is the feature-absorption correlation value, r H is the correlation threshold; the region with a feature-absorption correlation value of 100 is extracted as the region of the feature-absorption correlation map where the high-feature region and the key gamma-ray absorption region are closely correlated.

[0009] In a preferred embodiment, the image feature values and absorption feature values of the feature-absorption correlation map are extracted and superimposed on the corresponding region of the feature-absorption correlation map in the pathological section depth image to obtain the three-dimensional morphology map I of the reconstructed proliferative trichilemmoma pathological image F : ; Among them, E0(X, Y, Z) is the gray-scale feature value at the depth image (X, Y, Z) of the pathological section, DPA(X, Y, Z) is the image feature value at the feature-absorption correlation map (X, Y, Z), and SPT(X, Y, Z) is the absorption feature value at the feature-absorption correlation map (X, Y, Z). is the image feature weight coefficient. is the absorption feature weight coefficient.

[0010] Compared with the prior art, the present invention has the following beneficial technical effects: 1. By obtaining the pathological section image and its depth information to generate the depth image of the pathological section, the traditional two-dimensional pathological image can be extended to three-dimensional space, enabling doctors to not only observe the planar structure of tissues but also understand their characteristics in the depth direction, which helps to more comprehensively grasp information such as the morphology, size, and positional relationship of pathological tissues and reduce misjudgments caused by the limitations of two-dimensional observation.

[0011] 2. Emitting laser signals to detect pathological tissues and extract high-feature regions can prominently display parts of pathological tissues with special biological or morphological characteristics. These high-feature regions are related to the growth, differentiation, invasion, etc. of tumors, which helps doctors more accurately identify the lesion areas and provide a more precise basis for judgment.

[0012] 3. Using gamma rays to detect pathological tissues to generate gamma-ray absorption key regions can specifically highlight the tumor regions in pathological tissues. Gamma rays have high energy and penetration power and can penetrate tissues of a certain depth. By detecting their absorption, the boundary and scope of tumors can be clearly outlined, especially for deep tumors or tumors with low contrast with surrounding tissues, enabling more accurate positioning and evaluation.

[0013] 4. Combining the high-feature regions extracted by lasers and the absorption key regions generated by gamma rays and supplementing them into the three-dimensional information of the depth image of the pathological section, the resulting reconstructed three-dimensional morphology integrates various information, enabling doctors to more intuitively and comprehensively observe the pathological characteristics of proliferative trichilemmal tumors. This helps doctors make judgments quickly and accurately.

[0014] 5. The precise three-dimensional reconstruction image can provide detailed anatomical information for the formulation of subsequent treatment plans. Doctors can, based on the three-dimensional morphology, position of the tumor, and its relationship with surrounding tissues, through analyzing a large number of three-dimensional pathological images, deeply understand the pathogenesis and development process of proliferative trichilemmal tumors. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0016] Figure 1 Flowchart of the multi-modal ultrasound image fusion method based on artificial intelligence of the present invention; Figure 2 Digital pathological image of a trichilemmoma; Figure 3 Flowchart of extracting the original features of ultrasound images of the present invention; Figure 4 Schematic diagram of tissue absorption distribution image. Detailed implementation manners

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0018] First, obtain the pathological section image and the depth information of the pathological section image, and generate a pathological section depth image containing the depth information.

[0019] In this embodiment, the tissue submitted for examination is the surgically resected skin and subcutaneous tissue. The tumor is nodular and bulges above the skin surface, located in the dermis and subcutaneous tissue layers, and preferably has sizes of 2.0 cm × 1.5 cm × 1.0 cm, 1.2 cm × 1.0 cm × 1.0 cm, 1.8 cm × 1.5 cm × 1 cm; in this embodiment, the diseased tissue of a patient with proliferative trichilemmoma is selected. The patient is 44 years old, male, and has a nodule on the scalp that bulges above the skin surface with a diameter of about 1 cm.

[0020] Select the diseased tissue of the above-mentioned patient with proliferative trichilemmoma, and cut the tissue into thin slices with appropriate thickness according to the conventional pathological section preparation process, generally with a thickness of about 3-5 microns, for subsequent observation and imaging.

[0021] Obtain a high-resolution pathological section image and the depth information of the pathological section image. The depth information is a distance map captured by a depth structured light camera.

[0022] Specifically, a pathological section scanner is used to scan pathological sections to generate high-resolution digital pathological images, a microscope is used to take pictures of pathological sections, a high-resolution camera is used to obtain two-dimensional images of pathological sections, and a digital pathological image processing tool library is used to read and operate large-scale pathological section images.

[0023] In a preferred embodiment, data preprocessing is performed on the collected pathological section images.

[0024] Normalization processing: The input pathological section images are normalized to unify the data range.

[0025] Feature extraction: A convolutional neural network (such as ResNet50 or SEU-Net) is used to extract the features of pathological section images.

[0026] Downsampling: A sparse sampling algorithm is used to downsample the pathological section images to reduce the data volume and improve the processing efficiency.

[0027] As Figure 2 Shown is the digital pathological image of trichilemmoma after data preprocessing. Among the pathological features of some external trichilemmomas, the proliferation of interstitial fibrous connective tissue is a significant manifestation. This proliferation phenomenon is often particularly obvious in the central region of the tumor, presenting a special change process. As the tumor develops, in the central region, the fibrous connective tissue continuously proliferates, gradually becoming dense and disorderly arranged. This appearance of irregular pseudo-infiltrative growth is likely to cause certain interference to the diagnosis and requires careful discrimination. It is necessary to comprehensively judge through various pathological features to accurately distinguish trichilemmoma from other tumorous lesions with similar appearances.

[0028] Optical Coherence Tomography (OCT) technology is adopted. Based on the principle of light interference, by measuring the reflection and scattering of light in pathological tissues, the depth information of different positions of pathological sections is obtained, as Figure 3 shown.

[0029] After obtaining the pathological section images and depth information, the generator receives the pathological section images and depth information as inputs to generate a fused image; the discriminator judges the authenticity of the fused image. The generator and the discriminator continuously conduct adversarial training, and the generator learns to generate more realistic fused images, making it difficult for the discriminator to distinguish. Eventually, an ideal fusion result is obtained, that is, a pathological section depth image containing depth information is generated.

[0030] Secondly, a laser signal is emitted to the pathological tissue to detect the pathological tissue and extract the high-feature regions of the pathological tissue; a gamma ray is emitted to the pathological tissue to detect the pathological tissue and generate a gamma ray absorption key region that highlights the tumor region of the pathological tissue.

[0031] The pathological features of proliferative trichilemmoma are characterized by the hyperplasia of outer root sheath cells of hair follicles, the tumor cells are distributed in masses or cords, and the stromal components are special. Corresponding to the captured laser images, the areas with dense tumor cells and special stromal areas present high-feature areas in the transmitted laser images, and the key areas of gamma-ray absorption reflect the areas where tumor cells are metabolically active or tracers are enriched. Therefore, a laser signal needs to be emitted later to detect the pathological tissue, extract the high-feature areas of the pathological tissue, and then emit gamma rays again to detect the pathological tissue and generate the key areas of gamma-ray absorption that highlight the tumor areas of the pathological tissue.

[0032] When emitting a laser signal to detect the pathological tissue and collecting the laser transmission signal, the pathological tissue absorbs the laser energy to generate the transmission signal, and the transmission data at different angles are obtained through the detector.

[0033] Collect the transmission data at different transmission angles to obtain the transmitted laser image. Use the mapping function to map the transmission data to a two-dimensional plane, calculate the feature map of the transmitted laser image, count the frequencies of the appearance of each feature value, and extract the high-feature areas. The specific method is as follows: Let the transmission data collected by the detector at the transmission angle θ be P(θ, r), where r is the radial coordinate, and the transmitted laser image I is obtained. PA The formula is:

[0034] where r ， represents the radial mapping coordinate at the coordinate (x, y) in the image at the transmission angle θ.

[0035] h(r - r ， ) is the mapping function to improve the image resolution, and the filtering degree is controlled by adjusting the filtering parameter β.

[0036] Extract the high-feature areas based on threshold segmentation: Calculate the feature map of the transmitted laser image I PA and count the frequencies of the appearance of each feature value.

[0037] Traverse all the set thresholds T and calculate the difference value :

[0038] ; where is the proportion of the two types of features after segmentation according to the threshold T; is the average feature value of the two types of features, is the global feature value.

[0039] For all the thresholds T traversed, calculate and select the one that makes The maximum T is used as the segmentation threshold, and the segmentation method corresponding to this threshold can maximize the difference between the two types of features in the transmitted laser image I PA to reach the maximum.

[0040] Taking the segmentation threshold corresponding to this maximum difference value as the segmentation criterion, the region with a feature value greater than the segmentation threshold T is identified as the high-feature region and extracted. Because under this segmentation, the difference between the high-feature region and other regions is the most significant, which can better highlight the high-feature part in the transmitted laser image.

[0041] Next, gamma rays are emitted to detect the pathological tissue.

[0042] The detector records the projection data formed by gamma rays, and an iterative algorithm is used to convert the projection data into an image of tissue absorption distribution. For the image of tissue absorption distribution I PT the key gamma-ray absorption regions highlighting the tumor regions of the pathological tissue are extracted.

[0043] Specifically, an image of tissue absorption distribution is iteratively transformed from the projection data recorded by the detector for gamma rays. As Figure 4 shown, the high-frequency noise in the image of tissue absorption distribution is suppressed by three-dimensional Gaussian convolution to obtain a denoised image, thereby obtaining a high-quality image reflecting the tumor regions of the pathological tissue.

[0044] Let the projection data recorded by the detector for gamma rays be p m , and an iterative algorithm is used to convert the projection data into an image of tissue absorption distribution: ; where, represents the absorption value of the tissue absorption distribution image corresponding to voxel i at the k-th iteration; represents the absorption value of the tissue absorption distribution image corresponding to voxel i at the (k + 1)-th iteration.

[0045] A m,i is the system response matrix, which describes the contribution of voxel i to detector unit m. A m,i is the system response matrix, which describes the contribution of voxel i to detector unit m and reflects the probability that the rays emitted from voxel i are detected by detector unit m; R m,i is the normalization factor, which is used to normalize the relevant calculations to ensure the rationality and numerical stability of the calculation process; k is the number of iterations, and the image of tissue absorption distribution is gradually optimized through multiple iterations to make the image more accurately reflect the true situation of the tissue.

[0046] p m is the projection data recorded by the detector for gamma rays, m is the detector element index, representing the projection signal intensity received by the m-th detector unit.

[0047] This iterative algorithm utilizes the projection data p of the detector m , combined with the system response matrix A m,i and the normalization factor R m,i , continuously updates the absorption value of each voxel i in the tissue absorption distribution image I PT , and gradually reconstructs the tissue absorption distribution image I from the projection data PT .

[0048] Perform three-dimensional Gaussian convolution denoising on the tissue absorption distribution image I PT , and calculate the denoised absorption value corresponding to voxel i in the tissue absorption distribution image : ; where Ω is the spatial neighborhood, G is the Gaussian kernel function, where j - i represents the spatial position difference between voxel j and voxel i, and ρ is the smoothing parameter that controls the degree of smoothing. The larger ρ is, the wider the Gaussian kernel and the more obvious the smoothing effect.

[0049] Based on the threshold-based segmentation method, select the absorption threshold T H , and extract the key regions of gamma-ray absorption: ; The absorption threshold T H needs to be determined according to clinical criteria or statistical methods. All regions with a value of 1 form the key regions of gamma-ray absorption.

[0050] Finally, combine the high-feature regions extracted from the pathological tissue and the key regions of gamma-ray absorption, supplement the three-dimensional information of the pathological section depth image, and obtain the three-dimensional morphology of the reconstructed proliferative trichilemmoma pathological image.

[0051] Map the extracted high-feature regions and key regions of gamma-ray absorption to the spatial positions of the pathological sections, and label key structures such as the trichilemmoma cell hyperplasia area and the metabolically active area.

[0052] Based on the spatial position relationship of multiple sections, use a three-dimensional reconstruction algorithm. Let DPA q represent the image feature value of the q-th neighboring voxel in the high-feature region, and let SPT q represent the image feature value of the q-th neighboring voxel in the key region of gamma-ray absorption.

[0053] Calculate the correlation coefficient r between the high-feature region and the key absorption region DS : ; where Q is the total number of neighboring voxels, and q is the q-th neighboring voxel is the average value of the image feature values of all neighboring voxels in the high-feature region, is the average value of the image feature values of all neighboring voxels in the absorption key region.

[0054] Based on the correlation coefficient r DS Generate a feature-absorption correlation map: ; where I G is the feature-absorption correlation value, and r H is the correlation threshold; extract the region where the feature-absorption correlation value is 100, that is the corresponding region when, which is the region of the feature-absorption correlation map where the high-feature region and the gamma-ray absorption key region are closely related. Extract the image feature value and absorption feature value of this feature-absorption correlation map, and superimpose them on the region of the corresponding feature-absorption correlation map in the pathological section depth image, (the region where I G = 100), to obtain the three-dimensional morphology map I F of the reconstructed proliferative trichilemmoma pathological image: ; where E0(X,Y,Z) is the gray-scale feature value at (X,Y,Z) of the pathological section depth image, DPA(X,Y,Z) is the image feature value at (X,Y,Z) of the feature-absorption correlation map, and SPT(X,Y,Z) is the absorption feature value at (X,Y,Z) of the feature-absorption correlation map, is the image feature weight coefficient, is the absorption feature weight coefficient.

[0055] It should be noted that the feature-absorption correlation map is a partial regional map in the pathological section depth image, that is, the region of key concern in the pathological section depth image. Therefore, the position at (X,Y,Z) of the feature-absorption correlation map corresponds to the same position at (X,Y,Z) of the pathological section depth image.

[0056] The pathological section depth image can provide morphological information at the cellular and tissue levels, but lacks metabolic and functional information. Laser signals can reflect the optical absorption characteristics of tissues and are sensitive to blood vessel distribution, blood oxygen content, etc.; gamma-ray absorption data can show the metabolic activity of tissues. By fusing the data of the three, comprehensive judgments can be made from multiple perspectives of morphology, function, and metabolism, reducing misdiagnosis and missed diagnosis.

[0057] Organize the results of the quantitative evaluation indicators of each reconstruction method into the form of Table 1 for comparison, intuitively showing the differences in reconstruction accuracy among different methods.

[0058] Table 1 Differences in reconstruction accuracy among different methods

[0059] As shown in Table 1, the reconstruction method of the present invention combines laser signals with gamma-ray detection, can finely capture the characteristics of pathological tissues, has a high resolution (0.5 μm), a low measurement error rate (2%) for tumor regions and characteristic structures, and a small deviation from the true value (1 μm).

[0060] Traditional two-dimensional image-based reconstruction methods rely only on planar image reconstruction, making it difficult to accurately present three-dimensional structures, with a low resolution (50 μm), a high error rate (20%), and a large deviation (20 μm).

[0061] Three-dimensional reconstruction methods based on pure optical imaging are limited by the resolution of the optical system and have insufficient ability to distinguish fine features, with a resolution of 1 μm, an error rate of 5%, and a deviation of 5 μm.

[0062] Three-dimensional reconstruction methods based on low-dose CT have poor resolution of fine features of soft tissues, with a resolution of 3 μm, an error rate of 8%, and a deviation of 8 μm.

[0063] In one embodiment, a computer device is also provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0064] In one embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0065] In one embodiment, a computer program product or a computer program is provided. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to execute the steps in the above method embodiments.

[0066] Those skilled in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., and are not limited thereto. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logics, data processing logics based on quantum computing, etc., and are not limited thereto.

[0067] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0068] The above-described embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A three-dimensional reconstruction method for pathological image features of proliferative tricholoma, characterized in that: include: Acquire a pathological slice image and depth information of the pathological slice image, and generate a pathological slice depth image containing the depth information; Emit laser signals, detect pathological tissue, and extract high-feature areas of pathological tissue; Gamma rays are emitted to detect pathological tissues and generate key gamma-ray absorption areas that highlight the tumor areas of pathological tissues. The high-feature areas extracted from the pathological tissues and the key gamma-ray absorption areas are combined and added to the three-dimensional information of the depth image of the pathological slices to obtain the three-dimensional morphology of the reconstructed hyperplastic trichomonoma pathological image.

2. The method for three-dimensional reconstruction of pathological image features of proliferative tricholoma according to claim 1, characterized in that: When collecting laser signals, pathological tissue absorbs laser energy to generate transmission signals. The transmission data at different angles are acquired through the detector. The transmission data at different transmission angles are collected to obtain a transmission laser image. The transmission data is projected onto a two-dimensional plane using a mapping function. The feature map of the transmission laser image is calculated, the frequency of occurrence of each feature value is counted, and high feature areas are extracted.

3. The method for three-dimensional reconstruction of pathological image features of proliferative tricholoma according to claim 2, characterized in that: Assume that the transmission data collected by the detector at the transmission angle θ is P(θ, r), r is the radial coordinate, and the transmission laser image I is obtained. PA The formula is: ； Among them, h (rr ， ) is the mapping function, r ， Represents the radial mapping coordinates of the image coordinate (x, y) at the transmission angle θ; Calculate the transmitted laser image I PA The feature map is used to count the frequency of each feature value, and all the set thresholds T are traversed to calculate the difference value. : ； in, It is the proportion of the two types of features after segmentation according to the threshold T; is the average eigenvalue of the two types of features, is the global eigenvalue; Traverse all thresholds T and calculate , select The largest T is used as the segmentation threshold, and the areas with feature values ​​greater than the segmentation threshold are identified as high-feature areas and extracted.

4. The method for three-dimensional reconstruction of pathological image features of proliferative tricholoma according to claim 1, characterized in that: Assume that the projection data formed by the gamma ray recorded by the detector is p m ,The projection data are converted into tissue absorption distribution images using an iterative algorithm: ； in, represents the absorption value of the tissue absorption distribution image corresponding to voxel i at the kth iteration; represents the absorption value of the tissue absorption distribution image corresponding to voxel i at the k+1th iteration, A m,i is the system response matrix, describing the contribution of voxel i to detector unit m, R m,i is the normalization factor, k is the number of iterations, and the tissue absorption distribution image is gradually optimized through multiple iterations.

5. The method for three-dimensional reconstruction of pathological image features of proliferative tricholoma according to claim 4, characterized in that: Perform three-dimensional Gaussian convolution denoising on the tissue absorption distribution image and calculate the denoised absorption value corresponding to voxel i in the tissue absorption distribution image : ； Among them, Ω is the spatial neighborhood, G is the Gaussian kernel function, ji represents the spatial position difference between voxel j and voxel i, and ρ is the smoothing parameter that controls the degree of smoothing; Based on the threshold segmentation method, the absorption threshold T is selected H , extract the key areas of gamma-ray absorption: ； all The area with a value of 1 constitutes the key area of ​​gamma-ray absorption.

6. The method for three-dimensional reconstruction of pathological image features of proliferative tricholoma according to claim 1, characterized in that: Use DPA q Represents the image feature value of the qth neighboring voxel in the high feature area, using SPT q Represents the image feature value of the qth neighboring voxel of the gamma-ray absorption focus area, and calculates the correlation coefficient r between the high feature area and the absorption focus area DS : ； Among them, Q is the total number of neighborhood voxels, q is the qth neighborhood voxel, is the average of the image feature values ​​of all neighboring voxels in the high feature area, is the average value of the image feature values ​​of all neighboring voxels in the absorption focus area; Based on the correlation coefficient r DS Generate a feature-absorption correlation plot: ； Among them, I G is the characteristic-absorption correlation value, r H is the correlation threshold; the area with the extracted feature-absorption correlation value of 100 is the area of ​​the feature-absorption correlation diagram where the high feature area and the gamma-ray absorption key area are closely correlated.

7. The method for three-dimensional reconstruction of pathological image features of proliferative tricholoma according to claim 6, characterized in that: The image eigenvalues ​​and absorption eigenvalues ​​of the feature-absorption correlation map are extracted and superimposed on the region corresponding to the feature-absorption correlation map in the depth image of the pathological slice to obtain a reconstructed three-dimensional morphological map I of the pathological image of the proliferative trichomonoma. F : ； Among them, E0(X,Y,Z) is the grayscale eigenvalue at the depth image (X,Y,Z) of the pathological section, DPA(X,Y,Z) is the image eigenvalue at the feature-absorption association map (X,Y,Z), and SPT(X,Y,Z) is the absorption eigenvalue at the feature-absorption association map (X,Y,Z). is the image feature weight coefficient, is the absorption feature weight coefficient.

Images